Claude Cursor Skill

docker-project-foundations

Use this skill when setting up, initializing, or Dockerizing a project, even if the user doesn't explicitly mention Docker but describes a need for containerized local development, adding a database or cache dependency, or running services without host-level installs. Covers Dock

LLM Mart · 0 points · 3 views 0 listing impressions 0 install-command copies
Download docker-skills-skills_docker-project-foundations-a65ae5b.zip · 12 KB

Install

skills CLI npx skills add https://github.com/docker/skills/tree/main/skills/docker-project-foundations
Claude Code claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install docker-skills@llmmart
Git git clone https://github.com/docker/skills.git

The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole docker/skills collection as a plugin from our marketplace. Git is the plain clone.

Skill manifest

Docker Project Foundations

Overview

This skill guides you in Dockerizing a project from scratch. It focuses on creating the initial Docker file set, choosing a sane layout, and preferring containerized dependencies over host-level installs.

When to use this skill

Activate this skill when:

  • A user asks you to set up, initialize, or Dockerize a project
  • A project needs an initial Dockerfile, compose.yaml, or .dockerignore and does not have one
  • A user wants to add a service dependency (database, cache, message queue) to a project
  • A user asks how to run or develop a project locally and Docker is available

Do not use this skill when

Do not use this skill when:

  • The user explicitly wants to avoid Docker
  • The project already has a mature Docker setup and only needs minor edits
  • The main task is optimizing an existing Dockerfile
  • The main task is editing or debugging an existing Compose stack

Core guidance

Always create these three files

When Dockerizing a project, always produce all three:

  1. .dockerignore — Create this first so the initial build context is small and safe. See assets/dockerignore-example for a reference.
  2. Dockerfile — Create a working starter image definition that the project can build and run with. Start from assets/Dockerfile.simple and keep its baseline, even for a first scaffold:
    • multi-stage build (dependencies, build, and a separate runtime stage) whenever the project has a build or dependency-install step;
    • base image pinned to a specific version (e.g. node:22-slim), never latest;
    • dependency manifests (package.json + lockfile) copied before the source for layer caching;
    • a non-root USER set in the final stage before CMD/ENTRYPOINT.
  3. compose.yaml — Create a local development stack that includes the application service and any required dependencies. See assets/compose-dev.yaml.

npm registry credentials

  • Exclude .npmrc at every depth with **/.npmrc in .dockerignore; otherwise a broad source copy can persist credentials in image layers.
  • The Node.js starter mounts npmrc as a BuildKit secret for both npm ci steps. Public-package builds need no secret. For private registries, pass the config explicitly:
    DOCKER_BUILDKIT=1 docker build --secret id=npmrc,src="$HOME/.npmrc" .
    
    Use the actual config path if the project keeps it elsewhere. Never copy the credential file or pass its values through ARG or ENV. Dependency scripts run during installation can access the mounted secret; use trusted dependencies and a least-privilege registry token.
  • For private-registry builds through Compose, add this optional override as compose.npm.yaml alongside the starter's compose.yaml:
    services:
      app:
        build:
          secrets:
            - npmrc
    secrets:
      npmrc:
        file: ${NPMRC_PATH:?Set NPMRC_PATH to your npm config file}
    
    Build with NPMRC_PATH="$HOME/.npmrc" docker compose -f compose.yaml -f compose.npm.yaml build. This grants build-time access only, not a runtime secret. Public-package builds should omit the override so no credential file is required.

Prefer Dockerized dependencies over host installs

When a project needs a database (Postgres, MySQL, MongoDB), cache (Redis, Memcached), queue (RabbitMQ, Kafka), or any other infrastructure service:

  • Always define it as a service in compose.yaml instead of telling the user to install it on the host.
  • Never suggest brew install postgres, apt install redis, or similar host-level installs for development dependencies.
  • Use official Docker images from Docker Hub for these services.
  • Configure services with environment variables, not config files baked into images.

Bootstrap checklist

  • Name the file compose.yaml rather than legacy Compose filenames.
  • Put all three files at the project root unless there is a clear multi-service layout that justifies a docker/ subdirectory.
  • Ensure the initial setup can build and start locally with one command path.
  • Bind published application ports to loopback by default. Widen the host address only when another device must reach the development service.
  • Keep unauthenticated datastores on the Compose network instead of publishing their ports. If local host tools require database access, publish only to loopback.
  • If a development-only credential fallback enables one-command startup, label it clearly and document a .env override.
  • Use Compose services for local databases, caches, and queues instead of host installs.
  • Keep the first scaffold simple; defer detailed image optimization and advanced Compose tuning to the owning skills. "Simple" never means dropping the Dockerfile baseline above (multi-stage, pinned base, non-root USER).

Development vs production

  • Development: Use bind mounts for live reload, publish application ports on loopback by default, and enable verbose logging. Keep unauthenticated datastores on the Compose network; publish a datastore port only on loopback when local host tools require it.
  • Production: Use multi-stage builds, copy only built artifacts, do not mount source code, minimize image layers, set appropriate resource limits.
  • Keep a single Dockerfile that supports both via build stages and build arguments when possible.

File placement

  • Place Dockerfile at the project root (or in a docker/ subdirectory if the project has multiple services).
  • Place compose.yaml at the project root.
  • Place .dockerignore at the project root, next to the Dockerfile.

Related skills

  • For Dockerfile optimization, cache strategy, advanced non-root hardening, and image hardening, use docker-build-strategies. The starter baseline above (including a non-root USER) still applies here.
  • For service dependencies, health checks, overrides, volumes, networks, and Compose debugging, use docker-compose-patterns.
  • For destructive Docker CLI commands (docker system prune, docker rm -f, image/network/builder pruning) and a cross-product index of destructive-command guardrails, use docker-destructive-guardrails.

References

  • references/project-structure.md — Detailed guidance on Docker project file organization, naming conventions, and multi-service layouts.

Assets

  • assets/dockerignore-example — A comprehensive .dockerignore for a typical project.
  • assets/compose-dev.yaml — A development-oriented Compose file with Dockerized dependencies.
  • assets/Dockerfile.simple — A basic multi-stage Dockerfile following best practices.

Scripts

  • scripts/verify-setup.sh — Checks that required files exist in the current directory and validates its compose.yaml. Run it from the project root, with the script path resolved under this skill's directory:
    bash "<skill-dir>/scripts/verify-setup.sh" [--help]
    
    Replace <skill-dir> with the absolute path of the folder that contains this SKILL.md; the scripts/ path is relative to that folder, not to the project. Do not change into the skill directory first: the script checks the current directory. If the skill directory cannot be resolved, check that .dockerignore, Dockerfile, and compose.yaml exist in the project root, then run docker compose config --quiet. Exit status is 0 when verification succeeds or help is requested, 1 when required files are missing or the Compose configuration is invalid, and 2 for invalid arguments.

Checks

  • checks/verification.md — Detailed verification checklist for manual review.
Files (skills)
  • agents
    • openai.yaml 269 B
      Loading…
  • assets
    • compose-dev.yaml 1.5 KB
      Loading…
    • Dockerfile.simple 959 B · in bundle
    • dockerignore-example 654 B · in bundle
  • checks
    • verification.md 4.1 KB
      Loading…
  • evals
    • docker-project-foundations.md 6.8 KB
      Loading…
  • references
    • project-structure.md 3.3 KB
      Loading…
  • scripts
    • verify-setup.sh 1.1 KB
      Loading…
  • SKILL.md 7.9 KB
    ---
    name: docker-project-foundations
    description: Use this skill when setting up, initializing, or Dockerizing a project, even if the user doesn't explicitly mention Docker but describes a need for containerized local development, adding a database or cache dependency, or running services without host-level installs. Covers Dockerfile, compose.yaml, and .dockerignore creation with Docker best practices.
    license: Apache-2.0
    compatibility: Requires Docker 20.10+ and Docker Compose v2.
    ---
    
    # Docker Project Foundations
    
    ## Overview
    
    This skill guides you in Dockerizing a project from scratch. It focuses on creating the initial Docker file set, choosing a sane layout, and preferring containerized dependencies over host-level installs.
    
    ## When to use this skill
    
    Activate this skill when:
    
    - A user asks you to set up, initialize, or Dockerize a project
    - A project needs an initial `Dockerfile`, `compose.yaml`, or `.dockerignore` and does not have one
    - A user wants to add a service dependency (database, cache, message queue) to a project
    - A user asks how to run or develop a project locally and Docker is available
    
    ## Do not use this skill when
    
    Do not use this skill when:
    
    - The user explicitly wants to avoid Docker
    - The project already has a mature Docker setup and only needs minor edits
    - The main task is optimizing an existing `Dockerfile`
    - The main task is editing or debugging an existing Compose stack
    
    ## Core guidance
    
    ### Always create these three files
    
    When Dockerizing a project, always produce all three:
    
    1. **`.dockerignore`** — Create this first so the initial build context is small and safe. See `assets/dockerignore-example` for a reference.
    2. **`Dockerfile`** — Create a working starter image definition that the project can build and run with. Start from `assets/Dockerfile.simple` and keep its baseline, even for a first scaffold:
       - multi-stage build (dependencies, build, and a separate runtime stage) whenever the project has a build or dependency-install step;
       - base image pinned to a specific version (e.g. `node:22-slim`), never `latest`;
       - dependency manifests (`package.json` + lockfile) copied before the source for layer caching;
       - a non-root `USER` set in the final stage before `CMD`/`ENTRYPOINT`.
    3. **`compose.yaml`** — Create a local development stack that includes the application service and any required dependencies. See `assets/compose-dev.yaml`.
    
    ### npm registry credentials
    
    - Exclude `.npmrc` at every depth with `**/.npmrc` in `.dockerignore`; otherwise a broad source copy can persist credentials in image layers.
    - The Node.js starter mounts `npmrc` as a BuildKit secret for both `npm ci` steps. Public-package builds need no secret. For private registries, pass the config explicitly:
      ```bash
      DOCKER_BUILDKIT=1 docker build --secret id=npmrc,src="$HOME/.npmrc" .
      ```
      Use the actual config path if the project keeps it elsewhere. Never copy the credential file or pass its values through `ARG` or `ENV`. Dependency scripts run during installation can access the mounted secret; use trusted dependencies and a least-privilege registry token.
    - For private-registry builds through Compose, add this optional override as `compose.npm.yaml` alongside the starter's `compose.yaml`:
      ```yaml
      services:
        app:
          build:
            secrets:
              - npmrc
      secrets:
        npmrc:
          file: ${NPMRC_PATH:?Set NPMRC_PATH to your npm config file}
      ```
      Build with `NPMRC_PATH="$HOME/.npmrc" docker compose -f compose.yaml -f compose.npm.yaml build`. This grants build-time access only, not a runtime secret. Public-package builds should omit the override so no credential file is required.
    
    ### Prefer Dockerized dependencies over host installs
    
    When a project needs a database (Postgres, MySQL, MongoDB), cache (Redis, Memcached), queue (RabbitMQ, Kafka), or any other infrastructure service:
    
    - **Always** define it as a service in `compose.yaml` instead of telling the user to install it on the host.
    - **Never** suggest `brew install postgres`, `apt install redis`, or similar host-level installs for development dependencies.
    - Use official Docker images from Docker Hub for these services.
    - Configure services with environment variables, not config files baked into images.
    
    ### Bootstrap checklist
    
    - Name the file `compose.yaml` rather than legacy Compose filenames.
    - Put all three files at the project root unless there is a clear multi-service layout that justifies a `docker/` subdirectory.
    - Ensure the initial setup can build and start locally with one command path.
    - Bind published application ports to loopback by default. Widen the host address only when another device must reach the development service.
    - Keep unauthenticated datastores on the Compose network instead of publishing their ports. If local host tools require database access, publish only to loopback.
    - If a development-only credential fallback enables one-command startup, label it clearly and document a `.env` override.
    - Use Compose services for local databases, caches, and queues instead of host installs.
    - Keep the first scaffold simple; defer detailed image optimization and advanced Compose tuning to the owning skills. "Simple" never means dropping the Dockerfile baseline above (multi-stage, pinned base, non-root `USER`).
    
    ### Development vs production
    
    - Development: Use bind mounts for live reload, publish application ports on loopback by default, and enable verbose logging. Keep unauthenticated datastores on the Compose network; publish a datastore port only on loopback when local host tools require it.
    - Production: Use multi-stage builds, copy only built artifacts, do not mount source code, minimize image layers, set appropriate resource limits.
    - Keep a single `Dockerfile` that supports both via build stages and build arguments when possible.
    
    ### File placement
    
    - Place `Dockerfile` at the project root (or in a `docker/` subdirectory if the project has multiple services).
    - Place `compose.yaml` at the project root.
    - Place `.dockerignore` at the project root, next to the `Dockerfile`.
    
    ## Related skills
    
    - For Dockerfile optimization, cache strategy, advanced non-root hardening, and image hardening, use `docker-build-strategies`. The starter baseline above (including a non-root `USER`) still applies here.
    - For service dependencies, health checks, overrides, volumes, networks, and Compose debugging, use `docker-compose-patterns`.
    - For destructive Docker CLI commands (`docker system prune`, `docker rm -f`, image/network/builder pruning) and a cross-product index of destructive-command guardrails, use `docker-destructive-guardrails`.
    
    ## References
    
    - `references/project-structure.md` — Detailed guidance on Docker project file organization, naming conventions, and multi-service layouts.
    
    ## Assets
    
    - `assets/dockerignore-example` — A comprehensive `.dockerignore` for a typical project.
    - `assets/compose-dev.yaml` — A development-oriented Compose file with Dockerized dependencies.
    - `assets/Dockerfile.simple` — A basic multi-stage Dockerfile following best practices.
    
    ## Scripts
    
    - **`scripts/verify-setup.sh`** — Checks that required files exist in the current directory and validates its `compose.yaml`. Run it from the project root, with the script path resolved under this skill's directory:
      ```bash
      bash "<skill-dir>/scripts/verify-setup.sh" [--help]
      ```
      Replace `<skill-dir>` with the absolute path of the folder that contains this `SKILL.md`; the `scripts/` path is relative to that folder, not to the project. Do not change into the skill directory first: the script checks the current directory. If the skill directory cannot be resolved, check that `.dockerignore`, `Dockerfile`, and `compose.yaml` exist in the project root, then run `docker compose config --quiet`. Exit status is `0` when verification succeeds or help is requested, `1` when required files are missing or the Compose configuration is invalid, and `2` for invalid arguments.
    
    ## Checks
    
    - `checks/verification.md` — Detailed verification checklist for manual review.
    
  • skill.yaml 853 B
    Loading…

Comments (0)

Sign in to join the conversation.

No comments yet.

Reviews (0)

No reviews yet.